WO1997027469A1

WO1997027469A1 - Process and device for determining an analyte contained in a scattering matrix

Info

Publication number: WO1997027469A1
Application number: PCT/DE1997/000168
Authority: WO
Inventors: Alexander KNÜTTEL; Dirk Boecker; Matthias Essenpreis
Original assignee: Boehringer Mannheim Gmbh
Priority date: 1996-01-26
Filing date: 1997-01-23
Publication date: 1997-07-31
Also published as: CA2243571A1; ATE220202T1; EP0876596A1; KR19990076724A; IL125475A0; NZ330905A; AU2150397A; EP0876596B1; US5962852A; AU711422B2; DE59707650D1; ZA97599B; ES2177961T3

Abstract

A process is disclosed for determining an analyte in a scattering, in particular biological, matrix. In a detection step, detection measurements are made: light (primary light) is directed into the matrix (6) and light (as secondary light) emerging from the scattering matrix is detected to determine as a measured quantity a measurable physical characteristic of the light which can be altered by the interaction between light and matrix (6). In an evaluation step, data is obtained on the presence of the analyte in the matrix (6). The determination of optically weakly absorbing analytes against a strongly absorbing interfering background is improved by the use in combination of two selection processes for detecting the secondary light. In the first, the primary light (14) is focused by an optically focusing element (16) onto a focal region (17) located inside the matrix (6) at a specified measurement depth (d); the focal region (17) is focused by an optically focusing element (16, 21) onto a light inlet aperture (24) situated in the path of the secondary light (19) to the detector. In the second selection process, another depth-selection means, preferably a low-coherence measurement process, is used to detect light (secondary light) reflected from a defined measurement depth, the measurement depth corresponding to the focal depth.

Description

Method and device for determining an analyte in a scattering matrix

The invention relates to a method and a device for analyzing a scattering matrix with the aid of light on an analyte contained therein.

The most important application of the invention is the analytical examination of biological samples, in particular the tissue of a living organism. Biological samples are mostly optically heterogeneous, i.e. they contain a large number of scattering centers, at which incident light is scattered. This applies to human or animal tissue, in particular skin tissue or subcutaneous fat tissue, but also to liquid biological samples, such as blood, in which the blood cells form scattering centers or also to milk, in which the scattering centers are formed by emulsified fat droplets.

In addition, the invention is generally directed to scattering matrices in which an analyte is qualitative or quantitative should be determined. A scattering matrix in this sense is a three-dimensional structure with such a high density of optical scattering centers that penetrating light is generally scattered many times before it leaves the scattering matrix again. Non-biological scattering matrices that can be investigated on the basis of the present invention are, for example, emulsions and dispersions, as are required for different purposes, for example for lacquers and paints.

In the following, reference is made to the analysis of tissue as an example of biological and other scattering matrices without restriction of generality.

The aim of the analytical methods to which the invention is directed is to determine an analyte in the sense that information about the presence of a particular component contained in the tissue is obtained. The information can relate to the concentration of the analyte (quantitative analysis) or only to the question of whether the analyte (in a concentration above the detection limit of the method) is contained in the sample (qualitative analysis).

Up to now, analyzes of tissues and other biological samples have mainly been carried out invasively, i.e. a sample (usually a blood sample) is taken from the tissue, in which the analyte concentration is determined with the aid of reagents.

More recently, non-invasive analysis methods are increasingly being discussed, in which the analytical result is obtained painlessly and without reagents without taking samples from the tissue. Most of this Methods discussed for the purpose are based on the interaction of light with the scattering matrix. What these methods have in common is that measurement steps are carried out in which light is radiated into the matrix as primary light through an interface delimiting the matrix and light emerging from the matrix is detected as secondary light, by means of the interaction of the light with the matrix to measure variable, measurable physical properties of the light, which correlates with the concentration of the analyte in the matrix. Such a method step is referred to here as a "detection step" and the measurement as a "detection measurement", wherein a detection step can include one or more detection measurements.

The wavelengths of light that are discussed for such methods are generally between about 300 nm and several thousand nm, that is in the spectral range between the near UV and infrared light. The (detected) physical property of the light determined in the detection step, which can also be referred to as a "quantifiable parameter" (English: quantifiable parameter), is referred to below for the sake of simplicity as a "measured variable".

Since it is generally not possible to measure the analyte concentration in absolute terms in the methods discussed here, a calibration is required (as is the case with most analysis methods based on chemical reactions). Usually, in at least one calibration step, which is carried out in the same way as a detection step in terms of measurement technology, at least one detection measurement is carried out on a scattering matrix with a known analyte concentration. In the case of the analysis of living tissue, this is conveniently done by a comparative measurement with any known analysis method.

In an evaluation step of the analysis method, the analyte concentration is determined from the change in the measured variable in at least one detection step in comparison to at least one calibration step. The evaluation step comprises an evaluation algorithm in which the analyte concentration is determined in a predetermined manner from the results of at least one detection step and at least one calibration step.

Most methods of this type are based on the principles of spectroscopy, i.e. on the investigation of the spectral dependence of the optical absorption. For this purpose, detection measurements are carried out at at least two different light wavelengths. While this is a long-established problem-free method in a clear liquid, there are serious problems in the spectroscopic analysis of tissue and other scattering matrices.

On the one hand, the useful signal (the change in the absorption spectrum as a function of a change in the analyte concentration) is very small for most analytes, and this small useful signal is contrasted by a large background of interference signals, in particular from the optical absorption of water and others highly absorbent components (including the red blood dye hemoglobin) result.

Secondly, in a scattering matrix, due to the multiple scattering of the light, there is the problem that the optical path length covered by the light in the sample is unknown. However, knowledge of this path length is Prerequisite for being able to determine the concentration of an analyte in a spectroscopic analysis according to the Lambert-Beer law.

Numerous different attempts have been made to solve these problems. In particular, comparison techniques are used in which the attempt is made, the influences of highly absorbing interfering substances and also the influence of multiple scattering and the resulting lack of knowledge of the optical path length by means of several detection measurements and ratio or difference formation to eliminate. Time-resolved spectroscopy deals specifically with the problem of the unknown optical path length.

Despite these efforts, the spectroscopic analysis of tissue has only gained practical importance for one analyte, namely the red blood pigment hemoglobin (Hb or its oxidized form Hb0 ₂ ). These substances absorb so strongly and are present in such a high concentration that spectroscopic determination is possible using the known methods. However, it is precisely this strong optical absorption that is a major reason why the spectroscopic analysis of other analytes has so far been unsuccessful.

Glucose is a particularly important analyte because long-term successful therapy for diabetics requires continuous monitoring of the glucose level in the body. In order to avoid severe late damage, such as blindness or amputation of the limbs, the glucose concentration must be determined at least five times a day. This is hardly possible with invasive methods. However, the optical absorption of the tissue depends only to a very small extent on the glucose concentration. Therefore, spectroscopic principles have not led to success. Instead, various alternative methods for the non-invasive determination of the glucose concentration are discussed.

For example, in European patent specification 0074428 it is assumed that the glucose molecules characteristically scatter a light beam transmitted through a glucose solution and that the glucose concentration can be determined from the spatial angle distribution of the transmitted light intensity emerging from an examination cuvette or from an examined body part . In WO 94/10901 the spatial distribution of the secondary light intensity at the interface is determined as a measure of the glucose concentration and it is explained that this spatial distribution depends in a characteristic way on the glucose concentration in a tissue sample, because the glucose concentration due to the multiple scattering in the tissue influences the spatial distribution of the secondary light emerging from the interface to a surprisingly large extent. DE 4243142 A1 describes a method for determining the glucose concentration in the anterior chamber of the eye, the optical absorption and the rotation of polarized light being described as measured variables. Various possibilities are described in WO 95/30368 for determining the glucose concentration on the basis of LCI (Low Coherence Interferometry) measurements, the measured variables including the scattering coefficient and the refractive index, both of which are glucose depend on the concentration, are discussed. The methods described in these publications form important approaches for solving the problems associated with non-invasive analysis in tissue samples. However, they are limited to a specific analyte, namely glucose. In addition, the selectivity is relatively low due to the principle.

The invention is based on the object of providing a method and a device for analyzing tissue or other optically strongly scattering matrices, which makes it possible to selectively analyze an analyte contained therein even when total optical absorption of the scattering matrix is influenced only to a very small extent by the concentration of the analyte.

The object is achieved in a method which comprises at least one detection step and one evaluation step in the sense explained above in that two selection methods are used in combination with one another for the selective detection of secondary light coming from a defined measuring depth of the sample:

On the one hand, the primary light is focused by means of an optically focusing element, the primary light is focused on the focus area and the focus area is imaged by means of an optically focusing element on a light entry opening arranged in the light path of the secondary light to the detector, so that the detection of the Secondary light is concentrated on the focus area.

Secondly, an additional depth selection means is used to selectively detect reflected light as a secondary light from a defined measuring depth which corresponds to the depth of focus. A low-coherence reflectometric measuring method or a "time-gating" method is used in particular as an additional depth selection means.

The invention also relates to a device for carrying out such a method with a measuring head with a sample contact surface for application to an interface of the scattering matrix, light irradiation means with a light transmitter for irradiating primary light into the scattering matrix through the sample contact surface and the interface, detection means with a detector for detecting secondary light emerging from the scattering matrix through the interface and the sample contact surface, and evaluation means for determining the information about the presence of the analyte in the matrix from the measurement signal of the detector, the irradiation means and the Detection means each have an optically focusing element, the optically focusing element of the irradiation means focuses the primary light on a focus area located in the matrix at a depth of focus below the interface, and the optically focusing element Detection means maps the focus area onto a light entry opening arranged in the light path of the secondary light to the detector, as a result of which the detection of the secondary light is concentrated on the focus area. An additional depth selection means is provided in order to selectively detect light reflected from a defined measurement depth as secondary light, the measurement depth corresponding to the focus depth.

An arrangement in which an object is illuminated with light focused on a focal point (focus) and the observation is concentrated on the same focus referred to as "confocal arrangement". Such a confocal arrangement is known for various purposes, for example from the following quotes:

1) C.J.R. SHEPPARD et al. : "Imaging Performance of confocal fluorescence microscopes with finite-sized source", Journal of Modern Optics, 4_1 (1994), pp. 1521-1530.

2) U.S. Patent 5,192,980

3) U.S. Patent 5,345,306

4) EP 0689045

Since the light in the confocal arrangement is concentrated on the focus area at a certain depth of focus and also the detection concentrates on the same focus area, the confocal arrangement brings about a certain depth selection. The detector primarily detects photons which are reflected at a distance from the interface which corresponds to the depth of focus of the confocal arrangement. In order to achieve the advantageous results of the present invention, however, it is necessary to use an additional depth selection means, by means of which light which is selectively reflected from a defined measuring depth which corresponds to the depth of focus is detected as secondary light. As mentioned, a time-gating method is suitable for this. However, an additional depth selection using a low-coherence reflectometric measurement method is particularly preferred.

The confocal arrangement means that photons which are scattered on the way from the focus to the interface in the matrix and are thereby deflected cannot reach the detector through the light entry opening arranged in the light path in front of the detector. To the signal The detector therefore mainly contributes photons which are reflected on a structure in the area of the focus in the scattering matrix and which leave the matrix from there without being scattered.

In practice it is not possible to focus the primary light on a geometric point in the matrix and to focus the image on the same geometric point. Rather, optical aberrations and the finite size of the light source and the detector-side light entry opening inevitably lead to the partial volume of the matrix detected with the confocal arrangement having a finite size. This partial volume is referred to as the "focus area".

When using the method according to the invention and the corresponding device, changes in the concentration of an analyte, which has a very small share in the total optical absorption, lead to relatively large changes in the measurement signal. This enables selective analysis of such problematic analytes as well. This applies in particular in a wavelength range in which the analyte has an absorption band.

The invention is explained in more detail below on the basis of exemplary embodiments illustrated in the figures. Show it:

1 shows a schematic diagram of an analysis device according to the invention,

2a and 2b graphs of the dependence of the phase refractive index on the light wavelength for two different concentrations of a glucose solution in water, 3 shows a bar diagram of the group refractive index of glucose and of two interfering substances, each dissolved in water for four different wavelengths of light,

4 shows the differential group refractive index for the substances from FIG. 3 and two wavelength pairs each,

5 shows a part of an alternative embodiment of an analysis device in a basic sectional view,

6 shows a basic sectional illustration of a further embodiment,

7 shows a basic sectional illustration of a further embodiment,

8 and 9 show two principle sectional representations of different embodiments for realizing a spectrally resolved measurement at several light wavelengths,

10 shows a basic sectional illustration of an embodiment developed on the basis of FIG. 7.

The analytical device 1, which is highly schematic in FIG. 1, partly in section and partly as a block diagram, essentially consists of a measuring head 2 and an electronic unit 3. The measuring head 2 lies with a sample contact surface 4 at the interface 5 of a scattering matrix 6 (e.g. on the surface of human skin). In the measuring head 2 there are light irradiation means 8 for irradiating primary light 14 into the matrix 6 and detection means 9 for detecting secondary light 19 emerging from the matrix 6.

The light irradiation means 8 comprise a plurality of light transmitters 10, each of which has an optical system 8a (formed by the lenses 13 and 16 in the illustrated case) by which the primary light 14 is focused on a focus region 17 lying in the tissue at a predetermined depth d of the matrix. In the embodiment shown, a plurality of semiconductor light transmitters (light-emitting diodes) are monolithically integrated in a semiconductor substrate (chip) 11. In order to enable good focusing, the light transmitters 10 should be as point-shaped as possible. To limit the light exit openings 12, there are pinholes 12a in front of the light transmitters 10.

The diverging light emerging from the light exit openings 12 is collimated by an arrangement of collimation lenses 13 and enters a beam splitter cube 15 (beam splitter cube BSC) as a parallel beam. That from the opposite surface of the BSC

15 emerging light is through a focusing lens

16 focused on the focus area 17.

The light irradiation means, consisting of the light transmitters 10 and the components 12, 13 and 16, are present multiple times (n times) and are arranged as a (preferably regular) array 18 parallel to the interface 5 of the matrix 6 in such a way that primary light beams 14 are directed onto a A plurality of focus areas 17, which are preferably at the same measurement depth d in the tissue 6, are focused. For the sake of clarity, FIG. 1 shows the beam path only for one of the focus areas 17.

The secondary light emanating from one of the focus areas 17 is collimated by the lenses 16 and falls coaxially back into the primary beam. Beam splitting takes place in the BSC 15, the secondary light being reflected perpendicular to the direction of the primary light beam 14 in the direction of a detector 20. In front of the detector 20 there is a further lens 21 and a pinhole 22, which forms the detector-side light entry opening 24. The lenses 16 and 21 overall form an optical imaging system 9a, by means of which the focus region 17 is imaged on the light entry opening 24 (more precisely on the level of the detector-side light entry opening 24).

The detectors 20 are present in the same number as the light transmitters 10 n-fold and are arranged as an array corresponding to the light transmitters, with each detector 20 being assigned an optical system 16, 21. Semiconductor detectors (for example avalanche photodiodes) are preferably used which are monolithically integrated on a common semiconductor substrate 23. The collimator lenses 13, the focusing lenses 16 and the imaging lenses 21 arranged in front of the detectors 20 are each preferably designed as a "microlens array" (16a, 21a). Such microlens arrays are commercially available.

The optical systems 8a and 9a, which are a component of the light irradiation means 8 and the detection means 9, can be constructed in different ways in such a way that the explained lighting and imaging conditions are achieved. In particular, instead of the simple lenses shown, multi-lens systems (objectives), in principle also mirrors, can also be used as focusing elements.

A light focus in the focus area 17 in the scattering matrix is generated by the light irradiation means 8 using an optically focusing element (here the focusing lens 16). The confocal arrangement of the detection means 9 leads to (as above explained) reflected photons predominantly in the focus area 17 are detected by the detectors 20 and only a small proportion of diffusely scattered light reaches the detector 20.

The electronics unit 3 contains a power supply circuit 25 for the light transmitter 10, an amplifier circuit 26 for amplifying the output signals of the detectors 20 and an evaluation unit 27 which is usually realized with the aid of a microprocessor and which generates the desired analytical information from the Measurement signals of the detectors 20 are determined.

In order to provide an additional means for depth selection (ie for the selective detection of photons which were reflected in a measuring depth which corresponds to the depth of focus d) using a "time-gating" method, the power supply circuit generates 25 extremely short signal pulses, which are converted by the light transmitters 10 into very short light pulses. The amplifier circuit 26 and the evaluation unit 27 are set up to detect the secondary light detected by the detectors 20 within a defined time window, which corresponds to the desired measuring depth d. In view of the extremely short light-running times and the required precision, the means required for this are complex, but are practically available. A necessary control line for the transmission of a trigger signal is designated in FIG. 1 by 25a. Further details on different technologies suitable for such measurements can be found in the relevant literature. Reference is made in particular to the following publications, the content of which is also made by reference to the content of the present application text: 5) "Femtosecond optical ranging in biological systems", by JG Fujimoto et al. , OPTICS LETTERS, 1986, 150-152

6) "Time-resolved Fourier spectrum and imaging in highly scattering media", by L. Wang et al. , APPLIED OPTICS, 1993, 5043 ff ".

7) "A continuously variable frequency cross-correlation phase fluorometer with picosecond resolution", by

E. Gratton et al. , BIOPHYSICAL JOURNAL, 1983, 315-324

In the invention, several measurements with different focus depths d are preferably carried out. For this purpose, the measuring head 2 can be moved in the direction perpendicular to the interface 5. This vertical positioning can expediently be effected by means of a frame-shaped holding part 28 supported on the interface 5 of the scattering matrix 6 and a positioning drive which is symbolically represented in FIG. 1 as a double arrow 28a.

The combination of the metrological measures according to the invention leads to the fact that essentially only photons contribute to the measurement signal of the detectors 20, which were reflected in a defined measurement depth d and from there reach the respective detector 20 essentially without being scattered. Unscattered photons are also referred to as "ballistic" photons. However, the arrangement according to the invention also detects photons which are scattered with a small mean scattering angle and whose path (propagation path) runs in the vicinity of the geometric light path, which therefore deviate only minimally from the shortest (ballistic) path. Such photons are called "quasi ballistic". The invention The arrangement can therefore be described as a "depth-selective quasi-ballistic measuring regime".

According to the present knowledge of the inventors, the following reasons can be given for the advantages associated with this depth-selective quasi-ballistic measuring regime when determining an analyte in a scattering, in particular biological, matrix.

The ballistic or quasi-ballistic photons cover the shortest possible distance in the sample and are therefore absorbed relatively little. For this reason, optically strongly absorbing interfering substances contained in the sample (in the case of tissue samples, above all hemoglobin and water) interfere relatively little with the measurement. This effect also has to do with the fact that, despite the strong absorption of these interfering substances, the measuring light can penetrate relatively deep into the sample and, for example, layers can be reached in the skin in which the concentration of glucose changes significantly. Furthermore, the linear path of the photons detected in the method according to the invention means that the length of the optical light path in the sample is defined. It corresponds to twice the measuring depth d.

An effect which is explained in more detail below with reference to FIGS. 2 to 4 is of particular importance for the advantages obtained with the invention. Under the measurement conditions prevailing in the invention, the attenuation of the incident light intensity is determined predominantly by the scattering coefficient μ _s and only to a lesser extent by the absorption coefficient μ _a . It must be taken into account that, in contrast to a diffuse measurement regime, in which the depth-selective quasi-ballistic measurement range detected with the arrangement according to the invention gime the light transport is not described by the corrected scattering coefficient μ _s ', but by the uncorrected scattering coefficient μ _s , which is always larger than μ „. In the context of the invention it was found that the uncorrected scattering coefficient and thus the measurement signal under the measurement conditions according to the invention depend relatively strongly on the concentration of the analyte. Therefore, the analyte-specific signal change in the invention is relatively large.

FIGS. 2a and 2b show the course of the phase refractive index n ^ of a glucose solution in water at a glucose concentration of 6.25 mmol (millimoles per liter) or 100 mmol depending on the wavelength L in μm. These measurement results were obtained with an interferometer, with which the dependence of the phase on the Doppler frequency resulting from the movement of the interferometer mirror can be determined. The dependence of the phase refractive index on the wavelength shown in FIGS. 2a and 2b can be calculated directly from this. Of particular interest is the slope of the curve shown in the area of the absorption maximum at 2.15 μm (shown in broken lines in FIG. 2b). This slope corresponds to the group refractive index n _Q , which determines the scattering behavior in the scattering matrix.

In FIG. 3, the group refractive index n _Q for four different light wavelengths is plotted in the form of bar diagrams for three different substances, namely the analyte glucose (GLUC) and the two important interfering substances NaCl and lactate (LAC). FIG. 4 shows the differential refractive index, based on the wavelength. The group refractive index differences dn _{G are} plotted for the three substances and for the Wavelength pairs 1.6 μm minus 1.3 μm and 2.135 μm minus 1.75 μm.

The absolute values of the group refractive index Ώ shown in FIG. 3. _Q are highest for glucose, but also significant for the interfering substances NaCl and lactate. The differential refractive index shown in FIG. 4 shows a much clearer differentiation. For example, for the wavelength pair shown 2.135 μm minus 1.75 μm there is a value for dn _Q of 5 × 10 ⁶ / mmol. In this range there is an absorption maximum of glucose in water. From published literature data it can be seen that the differential absorption coefficient (ie the change in the absorption as a function of a change in the glucose concentration) dμ _a = 2 × 10 ⁴ / mmol (for a wavelength L = 2.15 nm).

In a scattering matrix, the light scattering, ie the scattering coefficient μ _g, depends on the refractive index, the size and the concentration of the scattering particles and on the refractive index of the medium in which the scattering particles are distributed. If these quantities are known, μ _s can be calculated using Mie theory, computer programs being available for such calculations. In the case of skin tissue as a scattering matrix, the following approximate values can be assumed:

Refractive index of the scattering bodies (cells): 1.41

Particle size: 10 μm

Concentration: 5%

Refractive index of the interstitial fluid: 1.38.

According to the Mie calculation, this results in a scattering coefficient of 6.8 mm ^-1 , ie one with measurement results corresponding value for the (uncorrected) scattering coefficient μ _s in tissue. This confirms that the assumed numerical values make sense. According to the Mie calculation, the above-mentioned value of the refractive index difference (differential refractive index) dn _G = 5xl0 ^"6 / mmol results in a corresponding differential scattering coefficient dμ _s = 2.015xl0 ^{~ 3} / mmol. This value is 10 times as high high as the mentioned differential change in the absorption coefficient (dμ _a = 2 × 10 ⁴ / mmol).

This shows that the measuring arrangement according to the invention detects a measured variable which is more sensitive to changes in the analyte concentration than the conventionally measured optical absorption in a diffuse measuring regime according to the prior art.

The measurement conditions and the method of evaluation can be optimized taking into account the knowledge according to the invention, the following considerations being taken into account.

Within the scope of the invention, at least two detection measurements with different light wavelengths are preferably carried out. In the evaluation step, a quotient of the measured values at the two different wavelengths is advantageously determined and the information about the presence of the analyte is determined on the basis of the quotient. A disturbing background absorption can thereby be largely eliminated if the total optical absorption of the sample changes much less than the scattering coefficient. This can be explained by the fact that the total attenuation of the light intensity as a function of the measuring depth d in the case of ballistic photons is proportional to e " ⁽ μs + μa ⁾ d -j_ _st - _ from this it follows that μ _a can be eliminated if a quotient is formed from the intensity measurement values of two measurements with constant μ _a .

In the case of skin tissue, medical concerns must be taken into account when choosing the measuring depth d. In order to detect a tissue layer whose glucose concentration is medically meaningful, the measuring depth should be at least about 0.3 mm. As described, larger measuring depths lead to an exponential decrease in the intensity of the measuring signal. A measuring depth of approximately 1.5 mm is currently regarded as the maximum upper limit on skin tissue.

At least two measurements are preferably carried out in which the measurement depth corresponding to the depth of focus d is different, with measurements with the same light wavelengths L ₁ and L _{2 being} carried out particularly preferably at the two different measurement depths d ₁ and d ₂ become. This advantageously makes it possible to avoid measurement errors which are attributable to fluctuations in the intensity of the incident primary light I _Q. Such fluctuations are not only to be feared as a result of intensity fluctuations in the light transmitter 10. In the case of measurements on the skin, the transition of the light from the measuring head 2 through the interface 5 into the sample 6 (“coupling”) is a fundamental problem. Even small changes in the position of the measuring head can cause changes in the intensity of the radiation entering the sample 6 Cause primary light that are larger than the measurement signal. These measuring errors can be eliminated in the measuring arrangement according to the invention if measurements are carried out with two different measuring depths and a quotient of the intensity signals measured is formed (in each case at the same wavelength). It is particularly advantageous if one of the two measurements is carried out with the smallest possible depth of measurement and the second measurement with the greatest possible depth of measurement. The small measuring depth should be below the epidermis layer. A range between 0.3 mm and 0.5 mm can be given as a guideline. The maximum size of the second (greater) measuring depth is essentially determined by the intensity of the measuring signal. The difference between the two measuring depths should be at least 0.3 mm.

The size of the light entry opening in front of the detector is also important. The term "true confocal imaging" was coined in the specialist literature in order to describe ideal conditions of a confocal imaging arrangement. More details on this can be found in the article by Sheppard et al. refer to. Within the scope of the invention, preference is given to working with a significantly larger light entry opening. Preferably, it should be at most five times the size required for true confocal imaging (according to the formulas given in the Sheppard et al. Article). Generally, the diameter of the light entry opening in front of the detector should be less than 0.1 mm, preferably less than 0.05 mm.

In practice, it is extremely advantageous not only to work with a single confocal arrangement and a single focus area, but also to use a large number of confocal arrangements in the form of an "array", through which the primary light is directed to a large number of focus areas in the scattering matrix focused and these focus areas are each imaged on a detector through an aperture. This increases the overall light intensity in medical Reached due to predetermined maximum power density. In addition, measurement errors due to micro heterogeneities in the skin are largely eliminated by averaging.

As mentioned, the desired analytical information is determined with the aid of an evaluation algorithm which links the measured values of the measured variable on the basis of a calibration with concentration values. In the present case, this linkage can consist of a simple one-dimensional evaluation curve which in each case assigns a concentration value to the quotient from the intensity measurement values of two wavelengths or two measurement depths.

Recently, more and more complex methods have been used in mathematics to improve the correlation between the measured variables (input variables) and the desired concentration (output variables) in analysis technology and thereby to achieve a higher accuracy of analysis . This includes, in particular, multilinear and nonlinear algorithms that link together several factors that are necessary for the evaluation of the analytical measurement. In the present case it can be expedient, for example, to carry out a large number of measurements at different light wavelengths in each measuring step and to correlate these measured values with the associated concentration value overall using a suitable numerical algorithm. Suitable algorithms are known and some are commercially available as computer programs. When using such a method, it can be advantageous to use the measurement signals as input variables for at least two different wavelengths and at least two different measurement depths directly (that is, without forming a quotient beforehand). Although the above explanations relate predominantly to glucose as analyte, the invention can also be used for other analytes for which the following conditions apply: low specific optical absorption of the analyte against a large absorption background; Total absorption of the sample is largely constant in the examined wavelength range; strong wavelength-dependent change in the optical absorption (and consequently also the refractive index of the interstitial fluid) in the examined wavelength range. Alcohol may be mentioned as a possible analyte with such properties. In individual cases, the suitability of the invention for the respective analysis can be tested by experimentally testing the arrangement according to the invention over a larger range of light wavelengths, and an optimal wavelength range can be determined.

FIG. 5 shows an embodiment in which a low-coherence reflectometric measurement is carried out as additional depth selection means in the measuring step. Such a method is also referred to as "LCI (low coherence interferometry) reflectometric measurement" or also as "optical coherence domain reflectometry (OCDR)". Such interferometric measuring methods are known for different purposes. Particular reference is made to the publication: "Measurement of Optical Properties of Biological Tissues by Low-Coherence Reflectometry", by Schmitt et al. , Applied Optics, 32nd (1993), 6032-6042 as well as WO 92/19930 and the already mentioned WO 95/30368.

For an LCI measurement, it is necessary that part of the light emitted by a spectrally broadband emitting light transmitter is divided by a beam splitter, and an optically reflecting one on a reference light path Element supplied, reflected by this and combined in front of the detector with the measuring light path in such a way that the secondary light and the reference light interfere with one another. The light receiver measures an interference signal if the optical light path length in the reference arm (from the beam splitter to the reflecting element) is a maximum of the coherence length of the light source from the optical path length of the measuring light path from the beam splitter to the point of reflection in the sample. An interference signal is only measured if this condition is met. This can be used to limit the examination of a sample to a certain measuring depth d.

In the embodiment according to FIG. 5, an additional light path is provided which is formed by light reflected to the left in the BSC 15 and which forms the reference arm 30 of an interferometer arrangement. The reference light is reflected on a movable mirror 31 and falls back into the BSC 15 in the opposite direction, it reaching the light inlet opening (pinhole 22) via the secondary light path 19. The interference condition is met if the optical light path in the reference arm 30 (up to the surface of the mirror 31) and in the sample arm 32 (up to the measuring depth d) match. The setting of the mirror 31 can thus improve the selective detection of the light reflected from the measuring depth d. Further details on the metrological implementation of the LCI method can be found in the aforementioned literature references.

Another special feature of the embodiment shown in FIG. 5 is that both the light irradiation means 8 and the detection means 9 each have optical fibers 33 and 34, respectively, by means of which the Light is guided from light transmitters, not shown, to the BSC or from there to detectors, also not shown. It is not of fundamental importance for the invention whether the primary light (as in FIG. 1) is radiated directly into the measuring arrangement by the light transmitters 10 or is detected directly by detectors 20 or whether light guides (as in FIG. 5) are used become. It is only essential that the effective light exit opening 12, from which the primary light enters the optical system and the effective light entry opening 24 in the secondary light path 19, have sufficiently small dimensions to ensure a sufficiently sharp focusing both in the primary light beam path and in the To enable secondary light beam path. The delimitation of the primary-side light exit opening and the secondary-side light entry opening can be realized in different ways, for example, as shown by pinholes, but also by the correspondingly dimensioned exit end of an optical fiber or by the size of the light-sensitive surface of a detector.

A problem with the arrangement with a BSC shown in FIGS. 1 and 4 is Fresnel reflections, which are caused by the refractive index jump at the interfaces of the BSC. As a result, light is reflected, for example, from the interface 29 of the BSC through which the measuring light exits into the sample (FIG. 5). This strong reflection signal can easily lead to an overload of the detector. In order to avoid this, in the embodiment shown in FIG. 6 an arrangement is selected in which the axis AP of the primary light beam 14 is inclined at an angle θ, which is less than 90 °, to the corresponding boundary surface 29. Thereby The light beam 35 specularly reflected by Fresnel reflection does not strike the detector 20.

The deviation of the angle θ is shown in a greatly exaggerated manner in FIG. In practice, a very small deviation of less than 1 ° is sufficient for the specularly reflected light no longer to strike the light entry opening 24 on the light detector side.

While in the embodiments of FIGS. 1, 5 and 6 the secondary light coaxially reflected back from the focus area is detected (ie the axis AP of the primary light and the optical axis AS of the secondary light between the optically focusing element and the focus point) FIG. 7 shows an embodiment in which the optically focusing elements of the irradiation means 8 and the detection means 9 are different, the optical axis on which the light is irradiated into the sample and the optical axis on which the light is detected to differ. In the embodiment shown, the primary light irradiated by an optical fiber 32 penetrates a first lens 40 and a second lens 41 for each focus area 17 of an array, which has a diameter twice as large as the first lens. The focused light enters the matrix 6 asymmetrically. After backscattering, the secondary light passes through the second lens 41 and a third lens 42 in the reverse order. The system of lenses 40, 41, 42 in this embodiment also ensures that the primary light 14 focuses on the focus area 17 and this focus area 17 the light inlet opening 24 on the detector side is imaged. The use of a total of three lenses with the arrangement shown has the advantage that commercially available microlens Arrays can be used, whereas in the case of a (basically also possible) oblique (ie not parallel to the surface 5 of the matrix 6) arrangement of the lenses, a special manufacture of the microlens arrangement is required.

In the embodiment according to FIG. 7, no troublesome problems with Fresnel back reflection can arise, since no light specularly reflected from a straight surface reaches the detector. Optical errors of the lenses (in particular aberration errors) lead to distortions and thus to a less sharp definition of the focus area 17. However, this is generally not a problem. If necessary, known optical corrective measures can be applied.

FIGS. 8 and 9 show two possibilities, in the case of an arrangement with the basic structure of FIG. 7, to enable measurement with a plurality of light wavelengths.

In FIG. 8, three laser diodes are used as primary light transmitters 10a, 10b, 10c, which emit light with different wavelengths and are expediently applied to a common substrate as integrated optics. These three light transmitters emit at different angles of incidence onto a half lens 45a, which collimates the light at different angles. Below the half lens 45a there is an optical grating 46, the grating constant of which is adapted so that the collimated beams of different light wavelengths incident at different angles are transformed into a common (collimated) beam along the optical axis. This is focused by a second half lens 45b onto a light exit opening 12. On the detector side there is a corresponding arrangement consisting of two half lenses 47a and 47b, the grating 46 and detectors 20a, 20b, 20c for the different light wavelengths. The two half lenses 45a, 45b, 47a, 47b form a sandwich with the grating 46, which can be constructed in a structurally simple manner with the aid of microlens structures.

In the embodiment according to FIG. 9, the three laser diodes are replaced by a single broadband laser diode 10. The wavelength selectivity is ensured here by the grating and the three detectors 20a, 20b, 20c arranged at different angles behind the grating. This arrangement is technically simpler with a lower wavelength selectivity than that of FIG. 8.

FIG. 10 shows the manner in which a low-coherence reflectometric measurement is possible as an additional depth selection means in an embodiment according to FIGS. 7 to 9. For the sake of clarity, only the central rays of the rays are shown. A horizontally running semitransparent mirror 48 is used here as a beam splitter to generate a reference beam. The reflected light beam 50a falls on a reference reflector 51, the distance of which from the semitransparent mirror 48 corresponds to the distance of the focus area 17 from the semitransparent mirror 48. For reasons of symmetry, the light striking the reference mirror 51 is focused, as is the light radiated into the sample 6. The light reflected by the reference reflector 51 falls back as a light beam 50b onto the semitransparent mirror 48 and is reflected from there in the direction of the reflector. The rays 50a and 50b form the reference arm 50 of the interferometer arrangement in this arrangement. A modulation of the reference light can be brought about by a vibrating reference reflector 51. Alternatively, an LCD element can be arranged in the light beam in such a way and controlled with a suitable modulation frequency that the reference light is modulated.

Claims

Expectations

Method for determining an analyte in a scattering, in particular biological, matrix, in which detection measurements are carried out in one detection step, in which light enters the matrix (6) as primary light through an interface delimiting the scattering matrix (6) irradiated and light emerging from the scattering matrix (6) through the boundary surface is detected as secondary light in order to determine a measurable physical properties of the light as a measured variable due to the interaction of the light with the matrix (6), and in an evaluation step, information about the presence of the analyte in the matrix (6) is determined on the basis of a measured value of the measured variable measured in the detection step, the primary light (14) using an optically focusing element (16; 40, 41) is focused on a focus area (17) lying in the matrix (6) at a depth of focus (d) below the interface, and the F Mapping the ocular area (17) by means of an optically focusing element (16, 21; 41, 42) to a light entry opening (24) arranged in the light path of the secondary light (19) to the detector, and thereby the detection of the secondary light (19) to the focus area ¬ rich (17) is concentrated, characterized in that by means of a depth selection means selectively reflected light from a defined measuring depth below the boundary surface (5) is detected as secondary light, the defined measuring depth corresponding to the focus depth (d).

2. The method according to claim 1, characterized in that at least two detection measurements are carried out with different light wavelengths.

3. The method according to claim 2, characterized in that in the evaluation step a quotient of the measured values at the two different wavelengths is determined and the information about the presence of the analyte is determined on the basis of the quotient.

4. The method according to any one of the preceding claims, characterized in that the primary light by means of an array (18) optically focusing elements (16;

40, 41) is focused on a plurality of focus areas (17) lying in the same focus depth (d) below the interface and the majority of the focus areas (17) by means of a corresponding number of optically focusing elements (16 , 21; 41, 42) is imaged on a respective light entry opening (24) arranged in the light path to the detector.

5. The method according to claim 4, characterized in that the primary light (14) is focused on at least ten different focus areas (17).

5. Method according to one of the preceding claims, characterized in that the one with the depth of focus corresponding measuring depth (d) is at least 0.3 mm and at most 1.5 mm.

7. The method according to any one of the preceding claims, characterized in that at least two detection measurements are carried out, in which the measuring depth corresponding to the depth of focus (d) is different.

8. The method according to claim 7, characterized in that the measuring depth corresponding to the depth of focus in the at least two detection measurements differs by at least 0.3 mm.

9. The method according to any one of the preceding claims, characterized in that a low-coherence reflectometric measurement is carried out as depth selection means in the measuring step.

10. The method according to any one of the preceding claims, characterized in that the primary light in the form of a short light pulse is radiated in as the depth selection means in the measuring step and the secondary light is measured within a defined time window.

11. The method according to any one of the preceding claims, characterized in that the analyte is glucose.

12. The method according to claim 11, characterized in that the wavelength of the light in at least one of the detection measurements is between 2.0 microns and 2.4 microns. 13. Device for carrying out the method according to one of the preceding claims with a measuring head (3) with a sample contact surface (4) for contacting an interface (5) of the scattering matrix (6),

Light irradiation means (8) with a light transmitter (10) for irradiating primary light (14) into the scattering matrix (6) through the sample contact surface (4) and the interface (5),

Detection means with a detector (20) for detecting secondary light (19) emerging from the scattering matrix (6) through the interface (5) and the sample contact surface (4)

Evaluation means (26, 27) for determining the information about the presence of the analyte in the matrix (6) from the measurement signal of the detector (20), the irradiation means (8) and the detection means (9) each having an optically focusing one Have element, the optically focusing element (13, 16; 40, 41) of the irradiation means (8) the primary light (14) in a focus area in the matrix (6) at a depth of focus (d) below the interface (5) (17) focuses and the optically focussing element of the detection means (9) maps the focus area onto a light entry opening (24) arranged in the light path of the secondary light (19) to the detector (20), whereby the detection of the Se ¬ secondary light (19) is concentrated on the focus area (17), characterized in that an additional depth selection means is provided in order to selectively detect reflected light from a defined measuring depth as secondary light, the depth of measurement coincides with the depth of focus (d).

14. The apparatus according to claim 13, characterized in that the diameter of the light entry opening is less than 0.1 mm.

15. The apparatus according to claim 13, characterized in that both the irradiation means (8) and the detection means (9) have an array of optically focusing elements through which the primary light (14) is focused on a plurality of focus areas (17) and Most of the focus areas (17) are imaged on a respective light entry opening (24) in front of a detector (20).

16. The apparatus according to claim 15, characterized in that the array of optically focusing elements is a microlens array (16a, 21a).

17. The apparatus of claim 15 or 16, characterized gekenn¬ characterized in that a plurality of detectors (20) on a semiconductor chip (23) is monolithically integrated.

18. Device according to one of claims 15 to 17, characterized in that a plurality of light transmitters (10) are monolithically integrated on a semiconductor chip (11).

19. Device according to one of claims 13 to 18, characterized in that the same optically focusing element (16) is both part of the irradiation means (8) and part of the detection means (9), so that the optical Axis (AP) of the primary light (17) and the optical axis (AS) of the secondary tes (19) between the optically focusing element (16) and the focus area (17) and a beam splitter is arranged on the side of the focusing element (16) facing away from the matrix (6), through which the secondary light ( 19) is branched off to the detector (20).

20. Device according to one of claims 13 to 18, characterized in that the irradiation means (8) and the detection means (9) have separate optically focusing elements (40 and 42), the optical axis (AP), on which the primary light (14) is radiated into the matrix (6) and the optical axis (AS) on which the secondary light (19) is detected are different.

21. Device according to one of claims 13 to 20, characterized in that it includes a low-coherence reflectometric measuring device as depth selection means.

22. Device according to one of claims 13 to 20, characterized in that it includes as a depth selection means a device to irradiate the primary light in the form of a short light pulse and to measure the secondary light within a defined time window.